Arithmetic geometry involves the interplay between algebra, number theory, and geometry to study spaces of solutions to systems of polynomial equations. Computers have an important role to play in exploring these spaces, whether it's numerically approximating solutions or counting solutions over finite fields. I have engaged in undergraduate research on topics such as: noncommutative algebras (especially Clifford algebras), error-correcting codes (especially those coming from algebra), enumerating orbit spaces of large group actions (especially those coming from polynomials), point counting and the Weil conjectures (for solving “finding hay in the haystack” problems), and cryptography. I am also interested in working on projects spanning the history of mathematics and in projects combining art and mathematics. Most projects would require some prerequisite background in algebra, such as Math 71. If algebra and number theory have piqued your interest and you are interested in learning more, come talk to me!

Professor Chernov advises students on various knot theory projects and the interactions of knot theory and general relativity. The knowledge of Math 54 Pointset Topology is required. The possible projects include Presidential Scholar research and Senior Theses.

I sometimes supervise undergraduate research projects in combinatorics, including senior theses, Presidential Scholars, WISP, and other independent projects. For most of the projects, having some background in combinatorics (such as math 28, 38 or 68) and algebra (24 and 31/71) is helpful, along with some programming skills. View some of the projects that I've been working on.

I am working on problems in an area of mathematics called homological mirror symmetry. The goal of this field is to understand a surprising relationship between two fields of mathematics that was first observed by string theorists. This duality predicts a correspondence between geometric shapes determined by polynomials (algebraic geometry) and the more flexible geometry of two-dimensional areas (symplectic geometry). In particular, mirror symmetry can point to new unexpected facts about combinatorial objects such as polytopes and hyperplane arrangements. I currently have a few problems in mind of this flavor.

To get started, students need at least to be comfortable with mathematical proofs and linear algebra, but additional background knowledge certainly would be useful. If you're interested but unsure if you're prepared, please reach out to discuss.

There are numerous opportunities to perform undergraduate research in my group on the mathematics of networks (how things are connected) and network representations of data. Previous students have developed new methods and written code for clustering networks into “communities” and other network measures, developed simulations for studying the spread of behaviors on networks, and studied network representations of a wide variety of application data sets from brain connectomes to political systems, and many other examples.

I have a number of projects accessible to undergraduate students in Combinatorics, Algebra and Graph Theory. These projects can lead to a senior thesis for honors or high honors. The ideal student should have taken Math 28 and Math 24 (or Math 22) if interested in a combinatorics project, Math 38 and Math 24 (or Math 22) if interested in a graph theory project, and Math 28 and Math 71 (or Math 31) if interested in an algebra project. Ideally the student should have some programming skills. For more details schedule an appointment. Here is a sample project with undergraduate student Geoffrey Scott.

Classical unsolved problems often serve as the genesis for the formulation of a rich and unified mathematical fabric. Diophantus of Alexandria first sought solutions to algebraic equations in integers almost two thousand years ago. For instance, he stated that if a two numbers can be written as the sum of squares, then their product is also a sum of two squares: since $5=2^2+1^2$ and $13=3^2+2^2$, then also $13\cdot 5=65$ can be written as the sum of two squares, indeed $65=8^2+1^2$. Equations in which only integer solutions are sought are now called Diophantine equations in his honor.

Diophantine equations may seem perfectly innocuous, but in fact within them can be found the deep and wonderously complex universe of number theory. Pierre de Fermat, a seventeenth century French lawyer and mathematician, famously wrote in his copy of Diophantus’ treatise “Arithmetica” that “it is impossible to separate a power higher than two into two like powers”, i.e., if $n>2$ then the equation $x^n+y^n=z^n$ has no solution in integers $x,y,z\ge 1$; provocatively, that he had “discovered a truly marvelous proof of this, which this margin is too narrow to contain.” This deceptively simple mathematical statement known as “Fermat’s last ‘theorem’” remained without proof until the pioneering work of Andrew Wiles, who in 1995 (building on the work of many others) used the full machinery of modern algebra to exhibit a complete proof. Over 300 years, attempts to prove there are no solutions to this innocent equation gave birth to some of the great riches of modern number theory.

Even before the work of Wiles, mathematicians recognized that geometric properties often govern the behavior of arithmetic objects. For example, Diophantus may have asked if there is a cube which is one more than a square, i.e., is there a solution in integers x,y to the equation $E : x^3-y^2=1$? This equation describes a curve in the plane called an elliptic curve, and a property of elliptic curves known as modularity was the central point in Wiles’s proof. One sees visibly the solution $(x,y)=(1,0)$ to the equation—but are there any others? What happens if 1 is replaced by 2 or another number?

Computational tools provide a means to test conjectures and can sometimes furnish partial solutions; for example, one can check in a fraction of a second on a desktop computer that there is no integral point on E other than $(1,0)$ with the coordinates x,y at most a million. Although this experiment does not furnish a proof, it is strongly suggestive. (Indeed, one can prove there are no other solutions following an argument of Leonhard Euler.) At the same time, theoretical advances fuel dramatic improvements in computation, allowing us to probe further into the Diophantine realm.

My research falls into this area of computational arithmetic geometry: I am concerned with algorithmic aspects of the problem of finding rational and integral solutions to polynomial equations, and I investigate the arithmetic of moduli spaces and elliptic curves. My work blends number theory with the explicit methods in algebra, analysis, and geometry in the exciting context of modern computation. This research is primarily theoretical, but it has potential applications in the areas of cryptography and coding theory. The foundation of modern cryptography relies upon the apparent difficulty of certain computational problems in number theory, in particular, the factorization of integers (in RSA) or the discrete logarithm problem (in elliptic curve cryptography).

I have several problems in the area of computational and explicit methods in number theory suitable for experimentation and possible resolution by motivated students. These problems can be tailored to the student based on interests, background, and personality, so there is little need to present the details here; but they all will feature a explicit mathematical approach and, very likely, some computational aspects. Mathematical maturity and curiosity is essential; some background (at the level of MATH 71) is desirable.